Ulcerative colitis (UC) is a chronic IBD, and its cause is unknown. It is confined to the mucosa and submucosa in the colon and characterized by repeated relapse and remission. UC can deteriorate the quality of life owing to relapsing symptoms including abdominal pain, bloody diarrhea, urgency, etc., and leading to complications such as colon cancer in the long-term.¹ Therefore, the management and treatment of this disease are important. Since the first incidence of UC was reported in the 1970s in Korea, its incidence and prevalence have increased continuously.^2,3 However, consistent management and examination are essential after diagnosis because, at present, there is no curative treatment for UC. Drugs used to treat UC include 5-aminosalicylic acid (5-ASA), steroids, immunosuppressants (azathioprine [AZP], 6-mercaptopurine [6-MP]), and others drugs. Patients with severe UC can be temporarily managed with corticosteroid injections.⁴ The association of infections with UC, such as infection with cytomegalovirus, needs to be examined in severe UC patients refractory to corticosteroids;^5,6,7 immunosuppressants including cyclosporine, infliximab, and other biological agents can be used concurrently.⁴ Cyclosporine tends to work more quickly than other immunosuppressive medicines. However, it needs to be used cautiously because of the risk of side effects such as nephrotoxicity, epilepsy, gingival hyperplasia, hypertrichosis, and other opportunistic infections, and it cannot be used in the long-term to maintain remission because of its potential side effects.^8,9,10 Infliximab is a monoclonal antibody against tumor necrosis factor alpha (TNF-α). It is known to alleviate histologic inflammation of the mucosa by reducing the secretion of mucosal TNF-α.^11,12 Two large-scale randomized controlled trials verified the effectiveness of infliximab in UC.¹³ However, infliximab is rarely used in patients with severe UC and information on its effects and adverse events are insufficient in Korea.

Therefore, this study aimed to examine the effects and adverse reactions of infliximab and identify predictive factors of response.

In this study, the use of infliximab clinically improved symptoms in almost 90% of moderate to severe UC patients refractory to steroids or immune modulators, showing a decrease in the baseline partial Mayo score by ≥3 points. Remission was reached in about 70% of all patients once or more than once. In addition, there was no death or surgery during the follow-up. Our results were higher than the response and remission induction rates reported in other studies performed in Korea.¹⁴ We assumed this was caused because remission was taken as standard criteria at any time point during the observation period, unlike other previous studies that determined clinical scores only at 8 weeks after the first infliximab administration. Moreover, this study aimed to identify predictive factors associated with the achievement of remission after infliximab administration. However, there was no clinical difference except for high Mayo scores indicating high disease activity at the first administration. Since these results were obtained from univariate analysis only, they were insufficient to be considered significant. Taking into consideration the fact that the CRP level was elevated and albumin level was low at the first administration in the remission group, high disease activity was predicted to have positive effect on remission induction caused by infliximab use. Likewise, a previous study that retrospectively reviewed 90 UC patients suggested that severe UC was a predictive factor for positive response to infliximab.¹⁵ However, high disease activity was considered a predictive factor for poor response to infliximab in several previous studies.^16,17,18 In contrast, a recent study involving 89 Korean UC patients proposed that a Mayo score of greater than 11 was a predictive factor for nonresponse to infliximab.¹⁸ Another recent study involving 134 Korean UC patients reported that CRP levels above 3.0 mg/dL and Hb levels above 11.5 g/dL were positive predictive factors for remission or partial response, and inconsistent results were shown for several factors representing disease activity in predicting the effects of infliximab.¹⁴ Therefore, additional studies are crucial to investigate causes for aggravation after remission from various perspectives of immunology, biochemistry, pharmacology, and environmental factors affecting disease progression.

Unlike the previous studies, this study analyzed "remission" only as the standard criteria instead of the remission state 8 weeks after infliximab use. Although remission is achieved before 8 weeks, relapse can occur after 8 weeks. Since the remission rate varies depending on total number and time of administration, as demonstrated by the results of the Kaplan-Meier analysis in this study and those of other large-scale studies, determining the effects and predictive factors of infliximab with "remission" achievement is thought to be more suitable rather than doing so with remission state at certain point of time. It is beneficial to compare remission state at specific point only if the aim of the study is to compare the remission rate between 2 groups, but the purpose of this study was to identify remission predictive factors. In the study, 3 patients reached remission and a short remission was followed by a relapse after the third administration (6 weeks after first administration). The partial Mayo scores for the patients were 5, 7, and 4 points, respectively. Those 3 patients accounted for 9.0% (3/33) of all subjects, and 13.0% (3/23) of all patients who achieved remission. According to the analysis based on 8 weeks of administration, an error can occur in the analysis of response-predictive factors by placing those 3 patients in the non-remission group. This is considered one of the reasons for our results not being comparable to those of previous studies on remission-predictive factors. Considering the fact that repeated remission and relapses can occur, whether remission at a specific point of time can be taken as the standard for identifying remission predictive factors instead of "remission" achievement needs to be carefully determined. To the best of the authors' knowledge, this issue has not been addressed in previous studies on the efficacy and response-predictive factors of infliximab in patients with UC, probably because remission achievement at a certain point of time was overlooked as massive data sets were analyzed in those studies. Although this study had a small sample size, new findings were revealed through in-depth analyses.

According to the Kaplan-Meier graph showing the remission rate of infliximab, the cumulative rate of remission maintenance decreased with time. However, factors related to remission maintenance rate were not identified. Many studies verified clinical improvement by administering infliximab to UC patients refractory to conventional therapies. The clinical improvement showed a decreasing tendency in both response and cumulative remission maintenance rates with time.^13,19,20 Some studies suggested that the decrease in the cumulative rates of remission maintenance was caused by a reduction in the plasma concentrations of infliximab, with the formation of antibodies against infliximab.²¹ In contrast, antibodies against infliximab present before drug administration were found to be not related to response to infliximab in UC patients.²² Furthermore, large-scale prospective study, a high response rate was observed at weeks 30 and 54 when antibodies were present in the bodies of UC patients.¹³ Therefore, additional studies are essential to further determine the antibodies and potential factors that cause non-response to infliximab. In this study, severe adverse events to the degree of following discontinuation of treatment occurred in 4 patients, but none of the patients died. Of these patients with side effects, in the case of 1 patient with pancreatitis, AZP was replaced with 6-MP, despite sustained use of infliximab. The adverse event was more likely to be caused by AZP. Therefore, relatively severe adverse reaction occurred in 3 (9.0%) out of 33 patients.

There are some limitations to this study. This study had a small sample size and relatively short follow-up period. Current health insurance coverage of infliximab for UC management is limited for those refractory or intolerant to conventional medications including steroid, AZP, or 6-MP and those with moderate to severe UC with a Mayo score of 6-12 points. Insurance coverage is discontinued for patients with no response to a maximum of 3 doses of infliximab. Therefore, most patients stop taking infliximab after 3 doses, if they are refractory to infliximab. Since response to the drug was mainly observed in the early period of administration in most previous studies, the small sample size is considered as a bigger barrier than the short follow-up period. However, the small sample size enabled in-depth analysis and helped reveal new findings. Moreover, there could have been differences between hospitals and physicians in treating disease owing to the nature of this multi-center study, so efforts were made to resolve this shortcoming by creating the objectified form of data input, considering the current condition of the actual clinical field at the data collection phase.

To sum up, this study analyzed a total of 33 Korean UC patients administered infliximab, and almost 90% showed clinical improvement. Some patients had a relapse followed by remission with time, and adverse events occurred in the case of some patients. It was found that the remitting and relapsing course can be repeated. In Korea, for moderate to severe UC patients refractory to conventional treatment with surgery as the last treatment option, infliximab is an effective and relatively safe treatment option. Further studies are required because the response to infliximab, potential side effects, and remitting and relapsing courses have not yet been fully clarified.